De Novo Cobalamin Biosynthesis, Transport, and Assimilation and Cobalamin-Mediated Regulation of Methionine Biosynthesis in Mycobacterium smegmatis

Abstract

Cobalamin is an essential cofactor in all domains of life, yet its biosynthesis is restricted to some bacteria and archaea. Mycobacterium smegmatis, an environmental saprophyte frequently used as surrogate for the obligate human pathogen M. tuberculosis, carries approximately 30 genes predicted to be involved in de novo cobalamin biosynthesis. M. smegmatis also encodes multiple cobalamin-dependent enzymes, including MetH, a methionine synthase that catalyzes the final reaction in methionine biosynthesis. In addition to metH, M. smegmatis possesses a cobalamin-independent methionine synthase, metE, suggesting that enzyme use-MetH versus MetE-is regulated by cobalamin availability. Consistent with this notion, we previously described a cobalamin-sensing riboswitch controlling metE expression in M. tuberculosis Here, we apply a targeted mass spectrometry-based approach to confirm de novo cobalamin biosynthesis in M. smegmatis during aerobic growth in vitro We also demonstrate that M. smegmatis can transport and assimilate exogenous cyanocobalamin (CNCbl; also known as vitamin B₁₂) and its precursor, dicyanocobinamide ([CN]₂Cbi). However, the uptake of CNCbl and (CN)₂Cbi in this organism is restricted and seems dependent on the conditional essentiality of the cobalamin-dependent methionine synthase. Using gene and protein expression analyses combined with single-cell growth kinetics and live-cell time-lapse microscopy, we show that transcription and translation of metE are strongly attenuated by endogenous cobalamin. These results support the inference that metH essentiality in M. smegmatis results from riboswitch-mediated repression of MetE expression. Moreover, differences observed in cobalamin-dependent metabolism between M. smegmatis and M. tuberculosis provide some insight into the selective pressures which might have shaped mycobacterial metabolism for pathogenicity.IMPORTANCE Alterations in cobalamin-dependent metabolism have marked the evolution of Mycobacterium tuberculosis into a human pathogen. However, the role(s) of cobalamin in mycobacterial physiology remains poorly understood. Using the nonpathogenic saprophyte M. smegmatis, we investigated the production of cobalamin, transport and assimilation of cobalamin precursors, and the role of cobalamin in regulating methionine biosynthesis. We confirm constitutive de novo cobalamin biosynthesis in M. smegmatis, in contrast with M. tuberculosis, which appears to lack de novo cobalamin biosynthetic capacity. We also show that uptake of cyanocobalamin (vitamin B₁₂) and its precursors is restricted in M. smegmatis, apparently depending on the cofactor requirements of the cobalamin-dependent methionine synthase. These observations establish M. smegmatis as an informative foil to elucidate key metabolic adaptations enabling mycobacterial pathogenicity.

Keywords: Mycobacterium tuberculosis; cobK; mycobacterial metabolism; riboswitch; tuberculosis; vitamin B12.

FIG 1

De novo cobalamin biosynthesis in M. smegmatis. (A) Cobalamin structure. The cobalt ion is coordinated equatorially by four nitrogen atoms of a corrin ring and axially by variable lower (α) and upper (β) ligands (R-group). Examples of β ligands are CN in cyanocobalamin (CNCbl; also known as vitamin B₁₂), adenosyl in adenosylcobalamin (AdoCbl; also known as coenzyme B₁₂), and methyl in methylcobalamin (MeCbl). The α ligand in the physiologically relevant cobalamin is typically dimethylbenzimidazole (DMB; outlined in red). (B) The final step in the methionine biosynthesis pathway is a nonreversible transfer of a methyl group from methyltetrahydrofolate (Me-THF) to homocysteine to produce methionine and tetrahydrofolate (THF). This reaction is catalyzed either by MetH using cobalamin as a cofactor or by MetE. MetE expression is attenuated by cobalamin via a cobalamin sensing riboswitch. (C) The LC-MS/MS method optimized to detect coeluting peaks corresponding to α-ribazole 5′-phosphate (highlighted in blue in panel A; blue trace in graphs) and DMB (red trace) transitions in a 20 ng/ml CNCbl standard. (D to F) Detection of de novo derivatized CNCbl. Cobalamin was detected in the wild type (D) but not in ΔcobK (E) and ΔmetE cobK::hyg (F) mutants. The wild-type and ΔcobK strains were grown in 7H9-OADC medium and the ΔmetE cobK::hyg strain was grown with 1 mM methionine supplementation. Peak intensities are expressed as counts per second (cps). (G) Growth curves of the ΔmetE cobK::hyg strain in liquid 7H9-OADC medium in the presence of 10 μM CNCbl or 1 mM methionine. The mutant cannot grow without supplementation.

FIG 2

Uptake of exogenous CNCbl and (CN)₂Cbi in M. smegmatis. (A) Spotting assays of exponential-phase cultures of wild-type and ΔmetE cobK::hyg strains on 7H10-OADC agar with or without 10 μM CNCbl or 10 μM (CN)₂Cbi show restricted uptake of (CN)₂Cbi relative to CNCbl uptake on solid medium. (B) alamarBlue assay to evaluate the growth of the ΔmetE cobK::hyg strain in liquid medium supplemented with (CN)₂Cbi. Cells (5 × 10³) were seeded in 7H9-OADC medium supplemented with 2-fold dilutions of (CN)₂Cbi starting at 30 μM as the highest concentration.

FIG 3

Assimilation of (CN)₂Cbi in M. smegmatis. (A) (CN)₂Cbi only control. (B) De novo-synthesized cobalamin in the wild-type strain. (C and D) Detection of recovered cobalamin due to (CN)₂Cbi assimilation in the ΔmetE ΔcobK::hyg double mutant. (E and F). Absence of recovered cobalamin in the ΔcobK strain in the presence of exogenous (CN)₂Cbi. (CN)₂Cbi uptake was accompanied by changes in the color of the spent medium from purple (A, inset) to a rusty hue in the wild-type (B, inset) and ΔcobK strains (F, inset), and pale yellow in the ΔmetE ΔcobK::hyg strain (D, inset). Supplemented cultures contained 30 μM (CN)₂Cbi.

FIG 4

Cobalamin-mediated attenuation of MetE expression in M. smegmatis. (A) ddPCR analysis of metE transcription during exponential growth phase in the wild-type and ΔcobK strains cultured in the presence (solid bars) or absence (open bars) of exogenous CNCbl. The ΔcobK strain exhibited an overabundance of metE transcripts relative to the wild-type parental strain. A small but statistically significant decrease in the level of metE transcript in the ΔcobK strain was observed in the presence of exogenous CNCbl (two-way ANOVA; *, P = 0.0359), but the change in metE transcript levels in the wild-type strain was not statistically significant. The graphed data are representative of two independent experiments. Error bars show the standard error of the mean. (B) Targeted MS analysis of MetE peptide levels (log₂ peak area) in the wild-type and ΔcobK strains grown in the presence (solid bars) or absence (open bars) of exogenous CNCbl. Exogenous CNCbl more significantly decreased MetE peptide levels in the wild-type strain relative to the ΔcobK mutant (two-way ANOVA; *, P = 0.0151).

FIG 5

Growth cessation due to methionine depletion in the metH cKD strain. (A) ATc-induced growth inhibition in the metH cKD strain is rescued by exogenous methionine (l-Met). (B) Representative images from time-lapse microscopy at the 0-h, 12-h, and 24-h time points showing severe growth retardation in the metH cKD strain. Images are taken from Movie S2, available at https://uct.figshare.com/s/65105b9914196c4b4654. Scale bars, 5 μm. (C) Quantification of microcolony growth in the M. smegmatis metH cKD strain using “R” software; *, limit of detection.

FIG 6

Sensitivity of MetH mutants to exogenous CNCbl. (A) CRISPRi-mediated silencing of metH in the ΔcobK strain in the presence (+) or absence (−) of exogenous CNCbl. Exogenous CNCbl had negligible effect on the growth of the metH cKD strain on solid medium in this background. (B) Spotting assay on 7H10 agar containing 10 μM exogenous CNCbl showing the failure of exogenous CNCbl to inhibit growth of the ΔcobK ΔmetH strain on solid medium. (C) Sensitivity of the ΔcobK ΔmetH strain to exogenous CNCbl in liquid medium, as determined using the alamarBlue assay.